Computers and Electronics in Agriculture 97 (2013) 71–82

Contents lists available at SciVerse ScienceDirect

Computers and Electronics in Agriculture

journal homepage: www.elsevier.com/locate/compag

Agricultural land use suitability analysis using GIS and AHP technique
Halil Akıncı ⇑, Aysße Yavuz Özalp 1, Bülent Turgut 2
Artvin Çoruh University, Faculty of Engineering, Department of Geomatics Engineering, 08000 Artvin, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study was to determine suitable lands for agricultural use in the Yusufeli district of Artvin
Received 16 December 2012 city (Turkey), where the current agricultural land in the district center and 3 villages will be completely
Received in revised form 1 July 2013 inundated while the land in 22 villages will be partially inundated due to three large dams currently
Accepted 6 July 2013
being constructed. The ‘‘Analytic Hierarchy Process (AHP)’’ method, commonly used in land use suitabil-
ity analysis, was utilized in this study. In application, the parameters of great soil group, land use capa-
bility class, land use capability sub-class, soil depth, slope, aspect, elevation, erosion degree and other soil
Keywords:
properties were used. In determining the weights of the parameters, experts’ opinions were consulted,
Geographical Information Systems
Multi-criteria decision analysis
and the agricultural land suitability map generated was divided into 5 categories according to the land
Analytic Hierarchy Process suitability classification of the United Nations Food and Agriculture Organization (FAO). After deducting
Agricultural land use suitability the forests, pastures and reservoir areas from the reclassified suitability map, it was estimated that 0.08%
Yusufeli of the study area (177.87 ha) is highly suitable for agricultural production, while 1.55% (3578.33 ha) is
moderately suitable and 6.3% (14575.91 ha) is marginally suitable for agricultural production. In addition,
it was found that the proportion of land that is currently unsuitable for agricultural production is 2.24%
(5183.63 ha), while the amount of land that is permanently unsuitable is 3.42% (7923.39 ha). It was also
determined that the following facts were all effective factors in reaching these results: a substantial por-
tion (approximately 85%) of the study area is covered with forests and pastures, the soil depth is inade-
quate for agricultural production, the slope in the study area is quite high and, accordingly, the erosion
degree is high.
Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction enable the planned and sustainable use of these resources in a

It is well-known that the rational and sustainable use of irrepro-
ducible natural resources such as land is one of the most important
indicators of economic growth. In fact, the World Commission on
Environment and Development associated land suitability with
sustainable development and defined sustainable development as
meeting the needs of the present without compromising the ability
of future generations to meet their own needs (Feizizadeh and
Blaschke, 2012). Moreover, rapid population growth and migration
require new areas in terms of meeting primary needs, particularly
in metropolises. In turn, these needs cause natural resources such
as forests, pastures, wetlands and agricultural land to be trans-
formed into settlements or industrial areas and cause these areas
to be used in ways that are unsuitable for their potential. For this
reason, it is extremely important to prepare land use plans that en-
able the transfer of natural resources to future generations and that
⇑ Corresponding author. Tel.: +90 466 215 10 00x1603; fax: +90 466 215 10 57.
E-mail addresses: hakinci@artvin.edu.tr (H. Akıncı), ayavuzozalp@artvin.edu.tr
(A.Y. Özalp), bturgut@artvin.edu.tr (B. Turgut).
1
Tel.: +90 466 215 10 00x1625; fax: +90 466 215 10 57.
2
Tel.: +90 466 215 10 00x1140; fax: +90 466 215 10 34.
manner that is suitable for their potential.
A prerequisite of land use planning is land suitability assess-
ment. The assessment in question guides towards the optimal
use of land by providing information on the opportunities and
constraints in the use of a given land area (Mokarram and Amin-
zadeh, 2010) and involves the decision to use available resources
based on their assessed potential (Bandyopadhyay et al., 2009).
For this assessment, the most suitable land use type is first
determined through land suitability analyses that consider land
properties and user needs (Akbulak, 2010; Amiri and Shariff,
2012). Land use suitability analysis is the process of determining
the suitability of a given land area for a certain type of use (agri-
culture, forest, recreation, etc.) and the level of suitability. An
important part of this process is the determination of the criteria
that affect the suitability of the land (Al-Shalabi et al., 2006). The
presence of various and multiple criteria makes land use suit-
ability analysis increasingly complex because, to support the
long-term use of a piece of land without deterioration, criteria
such as the socio-economic and environmental costs and conse-
quences must be taken into consideration in addition to the
inherent properties of that unit of land (Duc, 2006; Bandyopad-
hyay et al., 2009).

0168-1699/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.compag.2013.07.006
72 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82
Furthermore, it is understood that there is no certain standard
concerning the criteria to be taken into consideration when assess-
ing land suitability potential for agriculture and that the criteria
used in similar studies are usually those that are accessible. In
these types of studies, the topographical and soil properties of land
are widely used. For instance, Wang (1994) reported that agricul-
tural land suitability assessment involves the analysis of a large
variety and amount of physiographic data, including climatic char-
acteristics (temperature and rainfall), internal soil characteristics
(temperature, moisture, aeration, natural fertility, depth, texture
and salinity) and external soil characteristics (slope, flooding and
accessibility). In a study in which the physical and economic char-
acteristics of land were taken into consideration in land assess-
ment, Kalogirou (2002) used 17 physical characteristics classified
under three main factors as ‘‘soil mechanics and toxicities, slope,
erosion hazards, rooting conditions’’, ‘‘excess of salts’’ and ‘‘water
level, flood hazard and drainage’’. Perveen et al. (2007) used the
parameters of soil texture, soil moisture, soil consistency, soil pH,
soil drainage, organic matter content and slope in agricultural land
suitability analysis. While Zengin and Yılmaz (2008) included the
parameters of land use capability class, soil depth, limiting soil
properties, drainage, erosion, slope, aspect, presence of water, pre-
cipitation, temperature, vegetation cover and transportation in
assessing lands suitable for agriculture, Akbulak (2010) used the
parameters of slope, erosion, soil depth, limiting soil properties,
elevation and proximity to roads. To determine suitable land for
agriculture, Bandyopadhyay et al. (2009) used the parameters of
soil texture, organic matter content, soil depth, slope and land
use/land cover. Finally, in a land suitability analysis conducted in
Tabriz, Iran for agricultural production, Feizizadeh and Blaschke
(2012) used a total of 8 causal factors (elevation, slope, aspect, soil
fertility, soil pH, temperature, precipitation and groundwater) cat-
egorized within four main criteria they determined (namely,
topography, climate, soil properties and water resources).
As set forth above, to assess the suitability of a given piece of
land for agricultural production, many criteria need to be consid-
ered together. Because not all criteria affecting land suitability
have equal levels of significance, various methods were used in
determining the weights of these criteria and the scores of the
sub-criteria (Parakash, 2003). For instance, Perveen et al. (2007),
who aimed to determine physical suitability for rice crops, inte-
grated Geographical Information Systems (GIS) with the Multi-Cri-
teria Evaluation approach and used the Analytical Hierarchy
Process (AHP) method in application. In another study in which
GIS and AHP methods were integrated, Mustafa et al. (2011) as-
sessed the suitability of agricultural land in the Kheragarh Tehsil
area of India for several crops that can be cultivated in both sum-
mer and winter seasons. Similarly, in the study by Feizizadeh and
Blaschke (2012), who investigated the optimal use of land re-
sources for agricultural production, the authors carried out GIS-
based land suitability analysis using the AHP method. In addition,
Mokarram and Aminzadeh (2010) presented a GIS-based multi-cri-
teria land suitability evaluation using Ordered Weight Averaging
(OWA) with a fuzzy quantifier approach for agriculture. Mendas
and Delali (2012) developed a spatial decision support system en-
abling the preparation of agricultural land use suitability maps by
integrating GIS with the ELECTRE Tri (ELimitation Et Choix Tradu-
isant la REalité) multi-criteria analysis method.
The aim of the present study was to determine the areas suit-
able for agriculture in the Yusufeli district of Artvin city (Turkey)
using GIS and the AHP method. The AHP method is one of the mul-
ti-criteria decision-making approaches that are commonly used in
agricultural land use suitability analysis. The main purpose for
selecting Yusufeli district as the study area is the fact that the Art-
vin and Yusufeli Dams currently under construction, along with the
completed Deriner Dam, will either partially or completely inun-
date the fertile agricultural land in the district center and in 25 vil-
lages. Within the study area, in addition to agricultural land, the
district center along with the settlement areas of three villages will
also be inundated. Thus, work concerning the designation of new
settling areas for the settlements that will be inundated is being
carried out by the relevant institutions. The agricultural land suit-
ability analysis in this study was not carried out for a certain type
of crop production to prevent the land suitable for agriculture from
being selected as settlement areas and to enable the determination
of alternative areas for the agricultural land that will be lost be-
cause of the dams. This characteristic is one of the most important
differences that render this study, which generates considerable
results for the solution of an actual problem, valuable and distin-
guishes it from other similar studies in the literature. Product-
based (or crop-based) agricultural suitability analyses will be con-
ducted in future studies. Another contribution of the present study
is the importance of the main parameters used in terms of agricul-
tural production and the detailed explanation of the scoring ap-
proach used for the sub-parameters (how the upper and lower
limits of the parameters were determined) as distinct from other
studies in the literature. At the end of the study, the distribution
and areas of land suitable for agriculture in Yusufeli district were
determined.
2. Materials and methods
2.1. Study area
The study was carried out in the Yusufeli district, located south-
west of the Artvin city center (in Turkey). The study area is located
between the northern latitudes 40°340 20.0600 and 41°070 34.9100 and
the eastern longitudes 41°090 34.4100 and 41°540 32.2800 (Fig. 1). The
approximate surface area of the district is 2300 km2. According to
the 2011 data of the address-based population registration system,
the population of the district, which consists of one town and 62
villages, is 21,725. While 6318 people live in the district center,
15,407 live in the town and villages.
In the study area, the elevation varies between 320 m and
3937 m, and the elevation from sea level is approximately 600 m
in the district center. The average elevation in the village settle-
ments and agricultural areas is approximately 1300 m. However,
there are also villages and agricultural areas with elevations great-
er than 2000 m in the study area. In 89% of the district, the slope is
more than 30%, and there are scarcely any flatlands. According to
the long-term observation data (between 1970 and 2000) obtained
from the Turkish State Meteorological Service, the average annual
precipitation in the district is 278.5 mm, and the average monthly
temperature is 14.17 °C. The average lowest temperature is 1.1 °C
and occurs in January. The average highest temperature, however,
is 25.8 °C and is observed in July and August.
In many places in the study area where a rough topography is
dominant, the parent rock is visible due to the effect of the slope.
Alluvial soils formed by the accumulation of the materials carried
by streams are present on the streamside throughout the valley
(Koday and Erhan, 2010). According to the FAO soil classification
system, the soils of the region are included in the calcaric cambisol
class (Soil Atlas of Europe, 2005). Cambisols are young soils. Pedo-
genic processes are clearly visible in the color development and
structure of the sub-surface horizon. Cambisols correspond to the
inceptisol class in American soil classification.
Yusufeli district has a substantial amount of land, yet the area
suitable for sowing and planting is quite limited (Gök, 2010).
Rough and scattered village settlements where fruits and vegeta-
bles, particularly olives, rice and grapes, can be cultivated in lim-
ited areas are located in the valleys belonging to the Çoruh River
 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82 73

Fig. 1. Location map of the study area.

and its side branches. The most important cultigen that contributes
to the livelihoods of the local farmer families and trade in Yusufeli
and its surroundings is the olive. Also, greenhouse cultivation is
developing rapidly in the district and is considered to be an alter-
native source of income (Aytekin and Aktasß, 2010).
2.2. Parameters used in the land suitability analysis
humidity, natural efficiency, pH, salinity and organic substance
content was not needed. Third, the spatial data concerning the
parameters used had already been generated and made available
for use by the related institutions. The parameters used within
the scope of the study are explained in detail in the following
sections.

For the determination of the land in the study area that is suit-
able for agriculture, the parameters of great soil group (GSG), land
use capability class (LUCC), land use capability sub-class (LUCS),
soil depth, slope, elevation, erosion level and other soil properties
(OSP) were used. There are three main reasons for using these
parameters in this study. The first is the fact that the parameters
used in the study are adequate for determining the areas where
vegetative production can be carried out. The second reason is that,
because the study was not conducted for a certain type of agricul-
tural crop, the use of the parameters of precipitation, temperature,
2.2.1. Great soil group
Soil classification systems were created to aid the estimation of
the conditions with which the soils were formed and the behaviors
of the soils. Soil behaviors help in estimating the soil performance
for agricultural production. Thus, when deciding on the suitability
of land for agricultural production, it is necessary to know the
dominant soil class. The dominant soil group in the study area is
brown forest soil (Table 1). In the high elevation areas of the study
area, non-calcic brown forest soils and high mountain meadow
soils are available. Along the bed through which the Çoruh River
74 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82

Table 1 those that require special management. In the preparation of LUC-

Areal and percentile distributions of the main and sub-criteria parameters in the Cs, soils are grouped into eight classes (USDA, 2008) according to
study area.
the constraints they bring to the agricultural products to be culti-
Main Sub-criteria Area (ha) Area vated on them, the risk of loss of the crops in case the soils are used
criteria (%) for cultivation and the reactions the soils will give to management
GSG A (alluvial) 1275.87 0.56 applications. According to this classification, class I lands are plain
Y (high mountain meadow soils) 25879.59 11.18 or nearly plain lands that include deep and fertile soils that can be
M (brown forest soils) 135287.71 58.47
X (basaltic soils) 1596.29 0.69
easily processed by typical agricultural methods. Class II lands can
P (red–yellow podzolic soils) 442.86 0.19 be easily cultivated provided that certain special measures are ta-
N (non-calcic brown forest soils) 23624.03 10.21 ken. The difference between class I and class II fields is the avail-
CE (chestnut soils) 8.18 0.00 ability of one or more factors among the mild slope, medium
Water bodies and urban fabric 43284.47 18.70
level of erosion and the medium level of thick soil. Class III lands
LUCC I, II, III 1531.95 0.66 allow for agricultural production on the conditions that good plant
IV 4499.52 1.94
alternation is practiced and the proper methods are implemented.
VI 37835.69 16.35
VII 144247.38 62.34 Class IV lands are those that are particularly suitable to be consis-
VIII 42555.11 18.39 tently appropriated for pastures. Class V lands are not suitable for
No data 729.35 0.32 cultivated plants and are primarily designated for long-living
LUCS e (slope and erosion damage) 16261.51 7.03 plants such as pastures and forests. Class VI lands require a med-
es (slope and erosion damage, soil 134066.40 57.94 ium level of measures to be taken even for using the land for for-
inadequacy) ests or pastures. Class VII lands are those that are highly sloped
se (soil inadequacy, slope and erosion 10631.16 4.59
damage)
or exposed to a high level of erosion; those with stony, shallow
sw (soil inadequacy, wetness, inadequate 25973.31 11.22 or dry soils; swamps; or those with other inconvenient soil types.
drainage or flood losses) Lastly, class VIII lands contain factors that prevent them from being
w (wetness, inadequate drainage or flood 1182.15 0.52 used for agricultural production or as pastures or forestland
losses)
(TRGM, 2008). In addition to providing habitats for wildlife, these
Water bodies and urban fabric 43284.47 18.70
types of land can also be utilized as recreational areas. The preva-
Soil depth Deep (>90) 69.41 0.03
lent LUCC in the study area is class VII (Fig. 2b), followed in order
(cm)
Medium-deep (50–90) 17407.20 7.53 by class VIII and VI lands. Being unsuitable for agricultural produc-
Shallow (20–50) 58622.24 25.33 tion, these lands cover 97.08% of the study area. The area covered
Very shallow (0–20) 101880.97 44.03 by class I, II, III and IV lands was determined to be 2.6% (Table 1).
Litosolic 10134.71 4.38
Water bodies and urban fabric 43284.47 18.70
2.2.3. Land use capability sub-class
Erosion 1 (very week) 1275.87 0.55 Land use capability sub-class (LUCS) is the second category of
degree
land use capability class. Sub-classes represent the prevalent con-
2 (moderate) 44005.71 19.02
3 (severe) 132559.38 57.29 straints determining the capability class. Sub-capability classifica-
4 (very severe) 10273.57 4.44 tion is made based on the constraining factors in the land
Water bodies and urban fabric 43284.47 18.70 capability classification (TRGM, 2008). These sub-classes are ex-
OSP r (rocky) 109955.74 47.52 pressed through the use of the symbols (e) where the constraining
t (stony) 32410.50 14.01 factor is erosion or sensitivity to erosion; (w) for poor drainage, a
y (inadequate drainage) 1275.87 0.55
wetness problem, high ground water or flooding; (s) for root area
No data 44472.42 19.22
Water bodies and urban fabric 43284.47 18.70 constraints (e.g., soil shallowness, stoniness, high salinity, alkalin-
ity, low moisture holding capacity, infertility, etc.); and combina-
Slope (%) 0–2 107.94 0.05
2–6 735.41 0.32 tions of these symbols (e.g., es, se, sw, etc.) when more than one
6–12 2394.23 1.03 constraint exists and are given following the capability class (e.g.,
12–20 6533.07 2.82 I, IIe, IIIse, IVws, VIes, etc.). According to this method, in approxi-
20–30 14107.16 6.10 mately 80% of the study area, high slope and, accordingly, erosion
>30 207521.19 89.68
hazard and insufficient soil depth problems exist (Table 1). In addi-
Elevation 320–700 5807.50 2.51 tion, in areas close to the river and creek beds, drainage failure,
(m)
700–1000 20067.60 8.67
wetness and flood damage problems are observed (Fig. 2c).
1000–1300 29392.26 12.71
1300–1700 51557.20 22.28 2.2.4. Soil depth
1700–2100 48253.24 20.85 In general terms, soil depth indicates the depth to which the
>2100 76321.20 32.98
roots of cultivated plants can reach and use the available water
Aspect S, SW, SE 89649.78 38.74 and nutrients. In fact, water-storing capacity and effective rooting
N 26902.26 11.63
depth are related to soil depth. Soil depth is the most important
NW, NE 59146.78 25.56
W, E 55700.18 24.07 soil property affecting the hydrologic properties of soil and its
behaviors against erosion. Variation in soil depth is connected with
soil formation processes such as the chemical weathering of parent
rock, loss of material formed through weathering and transporta-
and Barhal Creek flow, alluvial soils formed by the accumulation of tion and deposition of soil by means of erosion (Fu et al., 2011).
the materials carried by the streams are available (Fig. 2a). In addition to soil formation and age, soil depth is also directly re-
lated to topography, parent material, living organisms and climate
(Gessler et al., 2000). The high levels of slope in the study area re-
2.2.2. Land use capability class sulted in low soil depth. As shown in Fig. 3a, in nearly 70% of the
Land use capability class (LUCC) indicates the suitability of a gi- study area, the soil depth is included in either the shallow or the
ven soil for the cultivation of many different cultigens except for very shallow class. The percentage of deep and medium-deep soils
 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82 75

Fig. 2. Great soil groups, land use capability classes and land use capability sub-classes maps used in the study.

Fig. 3. Soil depth, erosion and other soil properties maps used in the study.

that are more suitable for cultivation was determined to be 7.56%
(Table 1).
2.2.5. Erosion
Soil losses reduce soil fertility by negatively affecting the phys-
ical, chemical and biological properties of soils. Erosion reduces the
soil depth that is necessary for the development of plant roots and
the amount of water that the plants need, decreases the content of
nutritional elements and organic matter and consequently leads to
the formation of soil that is unsuitable for cultivation (Lobo et al.,
2005). Because soil formation processes are slower than weather-
ing processes, bare rocks that have no soil cover on top occupy a
significant place. In 57.29% of the study area, severe erosion is ob-
served (Fig. 3b). In 19.02% of the study area, moderate erosion is
observed, and in 4.44% of the area, a very severe erosion hazard
is observed (Table 1). In high elevations, the source of erosion is
rainfall; however, in lower elevations, the source of erosion is the
river and creeks.
2.2.7. Slope
The normal development of soils is closely related to the topog-
raphy of the area–in other words, with geomorphological proper-
ties. The thickness of the soil layer decreases with increasing
slope and increases with decreasing slope (Atalay, 2006). Slope de-
gree is the main factor determining erosion control (Koulouri and
Giourga, 2007). The amount of materials carried away with erosion
increases with the increasing degree of slope. Accordingly, with an
increase in slope degree, the development of soils occurs slowly
(Atalay, 2006), and soil depth and fertility decrease. On one hand,
slope indirectly limits agricultural production by affecting soil
properties negatively; on the other hand, slope negatively affects
agricultural production directly by restricting the possibility of
using machines and management applications such as soil tillage,
irrigation and drainage. Approximately 90% of the land in Yusufeli
district has more than 30% slope (Fig. 4c). That is, nearly the entire
area is included in the very steep and rugged class (Table 1).

2.2.6. Other soil properties 2.2.8. Aspect

Among the other soil properties that limit agricultural produc-
tion in the study area, rockiness is the most prominent (47.52%)
and is followed by stoniness (14.01%) and inadequate drainage
(0.55%) (Table 1). The rockiness and stoniness problems occurred
as a result of erosion damage. The inadequate drainage problem
was found primarily in areas close to river and creek beds. Fig. 3c
shows the OSP map of the study area.
To maintain their physiological activities, plants need sun expo-
sure at certain intervals. The duration of this need varies according
to the species of plant. However, in general, most cultigens exhibit
optimum growth in the southern and western aspects that receive
sunlight for a substantial portion of the day. For this reason, aspect
is taken into consideration as an assessment criterion for selecting
the land to be used for agriculture. It was determined that 62.78%
76 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82
Fig. 4. Topographical maps used in the study; (a) slope map, (b) aspect map, and (c) elevation map.
of the study area has no problem in terms of receiving sunlight (Ta-
ble 1). The aspect map of the study area is presented in Fig. 4b.
2.2.9. Elevation
Elevation is an important factor that plays a part in variation of
plant cover by causing temperature changes, particularly in high-
lands. While vertical temperature falls by 0.5 °C for every 100 m
within the first 4 km of the troposphere, it decreases by 0.6 °C be-
yond 4 km and by 0.7–0.8 °C close to the tropopause. Generally,
each 100 m of increase in elevation on mountains corresponds to
100 m of divergence from south to north or from lower latitudes
to higher latitudes. That is, with every 100 m of elevation on the
mountains, the vegetation periods and blooming of plants are de-
layed by 4–6 days (Atalay, 2006). This fact negatively affects the
plant variety to be selected for agricultural production. The lowest
elevation in the study area was found in the valley where the
Çoruh River flows. However, it was determined that the elevation
increases rapidly from the valley slopes and reaches values above
2100 m (Fig. 4a).
2.3. Analytic Hierarchy Process approach
The Analytic Hierarchy Process (AHP) is a multi-criteria deci-
sion-making approach introduced by Saaty (1977, 1994). The
AHP has attracted the interest of many researchers mainly because
of the effective mathematical properties of the method (Trianta-
phyllou and Mann, 1995). One of the most widely known and used
multi-criteria analysis approaches, the AHP method enables users
to determine the weights of the parameters in the solution of a
multi-criteria problem. In the AHP method, a hierarchical model
consisting of objectives, criteria, sub-criteria and alternatives is
used for every problem (Saaty, 1990). After the problem is set on
Table 2
The fundamental scale for pairwise comparison (Saaty, 1980).
Intensity of Definition
importance
1 Equal importance
3 Weak importance of one over another
5 Essential or strong importance
7 Demonstrated importance
9 Absolute importance
2, 4, 6, 8 Intermediate values between the two adjacent
judgments
Reciprocals If activity i has one of the above numbers assigned to it when compared with activity j, then j has the reciprocal value when compared with i
a hierarchical structure, the weights of the criteria forming the
hierarchy are calculated (Öztürk and Batuk, 2010). To evaluate
the criteria included in a level compared with other criteria in-
cluded in the next hierarchy level, scoring is made with the utiliza-
tion of the preference scale (Table 2) suggested by Saaty (1980),
and a pairwise comparison matrix is created (Saaty, 1980, 2004).
The pairwise comparison matrix consists of n(n  1)/2 compari-
sons for n number of elements (Malczewski, 1999; Öztürk and Ba-
tuk, 2010).
Solving a problem using AHP is carried out using the weights or
priorities of the criteria subjected to pairwise comparison. Weights
or priorities are determined by normalizing the pairwise compari-
son matrix. For this normalization, a ‘‘normalized pairwise com-
parison matrix’’ is obtained by dividing the column elements of
the matrix by the sum of each column. The row elements in the ob-
tained matrix are summed, and the total value is divided by the
number of elements in the row. In this manner, a priority vector
or weight vector is obtained (Tombusß, 2005). Weights are within
the range of 0–1, and their sum is equal to 1 (Malczewski, 1999;
Öztürk and Batuk, 2010).
While performing pairwise comparisons of criteria in the AHP
method, a certain level of inconsistency may occur. Therefore,
the logical consistency of pairwise comparisons must be checked
(Öztürk and Batuk, 2010). To measure the consistency of pairwise
comparison judgments, the consistency ratio proposed by Saaty
(1980) is used. A consistency ratio is calculated for the pairwise
comparison matrix. The upper limit proposed by Saaty for this ratio
is ‘‘0.10’’. In a case where the consistency ratio calculated for the
judgments is below 0.10, it is considered that the judgments
exhibit a sufficient degree of consistency and that the assessment
can be continued. If the consistency ratio is above 0.10, then the
judgments are considered inconsistent. In this case, the quality of
Explanation
Two activities contribute equally to the objective
Experience and judgment slightly favor one activity over another
Experience and judgment strongly favor one activity over another
An activity is strongly favored and its dominance demonstrated in practice
The evidence favoring one activity over another is of the highest possible order of
affirmation
When compromise is needed
 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82 77

the judgments needs to be improved. The consistency rate
can be reduced by reviewing the judgments (Öztürk and Batuk,
2007).
2.4. Data sets and methodology
The data concerning the topographical parameters used in this
study (slope, aspect and elevation) were obtained from 1/25,000
scale standard topographical maps in digital format. In application,
the Digital Elevation Model (DEM) of the study area was first cre-
ated using ArcGIS 9.3.1 software, and slope, aspect and elevation
maps were generated after the prepared DEM was converted into
the ESRI GRID format with a 25  25 m cell size.
The 1/25,000 scaled digital soil maps obtained from the General
Directorate of Agricultural Reform were used to obtain the data
concerning the characteristic properties of the soil in the study
area. The 25  25 m cell size ESRI GRID format GSG, LUCC, LUCS,
soil depth, erosion degree and OSP maps of the study area were
generated from ESRI Shape format soil maps by considering the
feature parameters in the national soil database (Fig. 5).
Data concerning present agricultural land and forest and pas-
tureland in the study area were obtained from the 1/25,000 scale

Fig. 5. Procedure followed in generating agricultural land use suitability map.

Table 3
Pairwise comparison matrix.

Criteria GSG LUCC LUCS Depth Slope Aspect Elevation Erosion OSP Weights
GSG 1 2 2 3 4 6 8 9 9 0.278
LUCC 1/2 1 2 3 4 6 7 9 9 0.233
LUCS 1/2 1/2 1 2 3 4 5 7 8 0.162
Depth 1/3 1/3 1/2 1 2 3 5 6 7 0.116
Slope 1/4 1/4 1/3 1/2 1 3 5 6 7 0.096
Aspect 1/6 1/6 1/4 1/3 1/3 1 1 2 3 0.040
Elevation 1/8 1/7 1/5 1/5 1/5 1 1 1 3 0.032
Erosion 1/9 1/9 1/7 1/6 1/6 1/2 1 1 2 0.025
OSP 1/9 1/9 1/8 1/7 1/7 1/3 1/3 1/2 1 0.018

Max. eigenvalue (kmax) = 9.7818.

n = 9.
Consistency index (CI) = (kmax  n)/(n  1) = 0.09773.
Random index (RI) = 1.45.
Consistency ratio (CR) = CI/RI = 0.06740.
78 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82

forest management plans in the ESRI Shape format, as obtained Table 4

from the Artvin Regional Directorate of Forestry. After obtaining Weights of the parameters and scores of the sub-parameters.

the coordinates and water elevation of Yusufeli, Artvin and Deriner Main Weight Sub-criteria Score
Dams from the Artvin Regional Directorate of State Hydraulic criteria
Works, reservoir areas were calculated using DEM. GSG 0.278 A (alluvial) 10
Afterwards, the pairwise comparison matrix was created to Y (high mountain meadow soils) 10
determine the weights of parameters according to the AHP M (brown forest soils) 8
X (basaltic soils) 7
method (Table 3). The judgments in the pairwise comparison P (red–yellow podzolic soils) 7
matrix (relative levels of importance of the parameters) were N (non-calcic brown forest soils) 5
determined by consulting the opinions of a team of experts, CE (chestnut soils) 3
including local agronomists and faculty members working in Water bodies and urban fabric 0
the Departments of Soil Science and Plant Nutrition in agricul- LUCC 0.233 I, II, III 10
ture faculties. The consistency ratio of the pairwise comparison IV 8
VI 2
judgments was calculated as 0.0674. Sub-parameters were also
VII 1
scored within the range of 0–10, again in line with experts’ opin- VIII 0
ions (Table 4). In this scoring, the diversity of the products that No data 0
can be cultivated in the area was taken into consideration. While LUCS 0.162 e (slope and erosion damage) 3
high points were given to the sub-parameters that positively af- es (slope and erosion damage, soil 2
fect the number of species that can be cultivated, lower points inadequacy)
were given to those that restrict the number of species that se (soil inadequacy, slope and erosion 2
damage)
can be cultivated. sw (soil inadequacy, wetness, inadequate 3
For instance, when evaluating the GSG sub-parameters, alluvial drainage or flood losses)
soils that constitute highly convenient areas for agricultural pro- w (wetness, inadequate drainage or flood 5
duction due to their physical and chemical properties were given losses)
Water bodies and urban fabric 0
10 points. Other soil groups, the physical and chemical properties
of which limit the diversity of the cultivable species, were scored Soil depth 0.116 Deep (>90) 10
(cm)
in light of the restricting factors. In this respect, the lowest points
Medium-deep (50–90) 8
were given to chestnut-colored soils because of their inadequacies Shallow (20–50) 6
in terms of soil depth, pH and organic matter content. Very Shallow (0–20) 2
Because agricultural production can be carried out on class I, II, Litosolic 1
and III land without the need of any preventative measures and be- Water bodies and urban fabric 0

cause of the high diversity of the species that can be cultivated in Slope (%) 0.096 0–2 10
these lands (as also stated in the LUCC definition), 10 points were 2–6 8
6–12 6
given to these lands. Intermediate scorings were made based on
12–20 4
the change in LUCC classes, and the class VIII lands that should 20–30 3
be left for wildlife were given 0 points. >30 1
When evaluating the study area in terms of LUCS, areas with Aspect 0.040 S, SW, SE 8
high levels of slope, severe erosion and soil insufficiency—causing W, EN 7
negative effects for agricultural production diversity—were given NW, NE 5
N 2
2 points. Moreover, areas exhibiting soil inadequacy and drainage
problems were given 3 points, whereas areas displaying drainage Elevation 0.032 320–700 10
(m)
problems were given 5 points.
700–1000 9
Soil depth is an important parameter in terms of plant root 1000–1300 7
development. The diversity of cultivable species increases based 1300–1700 6
on the depth of the plant root growth environment. For this reason, 1700–2100 4
when scoring the soil depth sub-parameter, soils deeper than >2100 2

90 cm, which are expected to increase the diversity of the cultiva- Erosion 0.025 1 (very weak) 10
ble species, were given 10 points, while the soils where the parent degree
2 (moderate) 8
rock is visible, soil formation is slow and, accordingly, the diversity 3 (severe) 6
of cultivable species is limited, were scored with 1 point. In line 4 (very severe) 4
with this consideration, intermediate points were given depending Water bodies and urban fabric 0
on soil depth. OSP 0.018 y (inadequate drainage) 6
Slopes directly affect the diversity of plant species because of t (stony) 4
their effects on mechanization during agricultural production and r (rocky) 2
No data 0
the problems they cause in terms of irrigation and drainage. Be-
Water bodies and urban fabric 0
cause such restrictions occur at the lowest possible rate on plain
and nearly plain areas, such areas were scored with 10 points.
However, because the number of cultivable species is highly lim-
ited in steep lands that have more than 30% slope, only 1 point
was given to such areas. The diversity of the species to be cultivated in a geographical
It was considered that the areas with southern, southwestern or position also varies according to the elevation from the sea. In par-
southeastern aspects allow for the cultivation of more products ticular, plants’ needs for temperature and both late spring frosts
due to the duration in which they receive light; therefore, such and late autumn frosts significantly affect the diversity of plant
areas were scored with 8 points. In contrast, because receiving species. Plant production diversity in the study area varies accord-
inadequate daylight is a factor that limits the diversity of cultivable ing to the elevation. It was determined that, at elevations between
species, areas with a northern aspect were scored with 2 points. 320 and 700 m, a large variety of vegetables and fruits, particularly
 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82
olives, can be cultivated; therefore, areas at these elevations were
given 10 points. In addition, it was determined that product diver-
sity decreases at higher elevations; at 2100 m elevation, it is not
possible to cultivate economic and agricultural products other than
grains and fodder crops. Therefore, these areas were scored with 2
points.
The diversity of the products that can be cultivated in areas sub-
jected to erosion falls because of the negative effects on the phys-
ical and chemical properties of the soils that result from the
decreasing soil depth. Therefore, areas subjected to severe erosion
were scored with 4 points, while areas exposed to the least amount
of erosion were given 10 points. Similarly, intermediate points
were given to the areas in line with the respective levels of erosion.
In addition to the parameters examined above, other soil prop-
erties also affect the diversity of the products that can be cultivated
in a given area. Because rocky areas limit the diversity of cultivable
products with their high level of slope and inadequate soil depth,
these areas were scored with 2 points. In addition, while 4 points
were given to stony areas, areas with inadequate drainage prob-
lems were scored with 6 points.
After parameter weights and sub-parameter scores were ap-
pointed to the related layers in the ArcGIS 9.3.1 environment, ras-
ter maps of 9 parameters were overlaid using the weighted sum
overlay analysis, and an agricultural land suitability map was gen-
erated. The analysis layer (agricultural land use suitability map)
was divided into 5 classes of equal ranges according to the land
suitability classification of FAO (1976) (Fig. 6). The agricultural
land use suitability map of Yusufeli district was obtained by
removing the forest, pasture and reservoir areas from the reclassi-
fied suitability map. The steps followed in this study are presented
in general terms in Fig. 5.
Fig. 6. Agricultural land use suitability map of Yusufeli district.
79
3. Results and discussion
According to the agricultural land use suitability map gener-
ated, it was determined that, while 18.7% (43287.03 ha) of the
study area is permanently unsuitable for agricultural production,
27.4% (63393.32 ha) is currently unsuitable, 51.2% (118478.51 ha)
is marginally suitable, 2.3% (5250.89 ha) is moderately suitable,
and 0.4% (989.25 ha) is highly suitable (Fig. 6).
High slope and erosion degrees along with the shallow soil
depth of the study area were effective factors, resulting in a low
rate of highly suitable agricultural land in this study. Similarly, in
_
a study conducted to evaluate agricultural land suitability for Ispir,
Erzurum (Turkey), parallel results were obtained, as a rate of 0.9%
for highly suitable agricultural land was determined (Demir et al.,
2011). The authors indicated that the very high mean slope values
_
in Ispir not only determine the general characteristics of agricul-
tural areas but also limit potential agricultural land in the region.
Forests and pastureland cover 85.2% (197187.90 ha) of the
study area (Table 5). While 87.5% (103661 ha) of the land that is
marginally suitable for agricultural use coincides with forest and
pastureland, 27.8% (1457.58 ha) of moderately suitable land and
15.8% (156.07 ha) of highly suitable land coincide with forest and
pastureland. Because agricultural activities cannot be carried out
in forest and pastureland, pursuant to legal regulations in Turkey,
these areas were removed from the suitability map, and a new
suitability map was obtained, as shown in Fig. 7.
With the consideration of the reservoir areas of the Yusufeli,
Artvin and Deriner Dams in the study area, it was calculated that
2771.97 ha of land will be inundated (Table 5). It was determined
that the land highly suitable for agricultural use will be particularly
affected by this inundation: 66.2% (655.31 ha) of this land will be
80 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82

Table 5
The distribution of agricultural land suitability analysis results based on area and percentage.

Suitability degree Total area classified by the Area within forest Areas to be inundated Suitable for agricultural
suitability analysis and pasture lands by the reservoirs production
ha % ha ha ha %
Highly suitable 989.25 0.4 156.07 655.31 177.87 0.08
Moderately suitable 5250.89 2.3 1457.58 214.98 3578.33 1.55
Marginally suitable 118478.51 51.2 103661.00 241.60 14575.91 6.30
Currently not suitable 63393.32 27.4 58020.01 189.68 5183.63 2.24
Permanently not suitable 43287.03 18.7 33893.24 1470.40 7923.39 3.42
Total 231399.00 100 197187.90 2771.97 31439.13 13.59

Fig. 7. Suitability map obtained after removing pasture and forest lands.

lost because of inundation, and their rates in the study area will fall
from 0.4% to 0.08%. Fig. 8 presents the final agricultural use suit-
ability map obtained by removing the reservoir areas of Yusufeli,
Artvin and Deriner Dams.
In addition, the extent to which the present agricultural lands
correspond with the suitability map and how they will be affected
by the dams were determined in this study. Examining the forest
management plan (forest stand map) of the study area shows that
only 8.2% (18995.74 ha) of the area is currently used for agricul-
tural activities. It was also determined that 61.95% (9034.43) of
the marginally suitable areas, 18.69% (670.31 ha) of the moderately
suitable areas and 0.91% (1.62 ha) of the highly suitable areas are
presently used for agricultural activities.
Currently, the building of large dams appears to be inevitable
considering that dams provide irrigation for agricultural produc-
tion, drinking water and hydroelectric power for ever-increasing
population growth. Consequently, building large dams covers an
important portion of Turkey’s investment budget (Yalçın and Eken,
2006). However, it is also well-known that large dams can severely
change the river systems they are built on and their immediate
environment (Toker, 2010). Large dams can cause social, cultural
and economic losses as they inundate settlements, agricultural
lands and cultural heritage. Similarly, in this study, it was deter-
mined that, besides the extensive settlements, including the dis-
trict center and three villages in Yusufeli town, 5.81%
(1103.86 ha) and 0.33% (62.68 ha) of the present agricultural lands
will be inundated with the reservoirs of Yusufeli Dam and Artvin
Dam, respectively. Various studies have revealed that several large
dams—both under construction and/or planned—within the Çoruh
Watershed have been causing negative effects on social, cultural
and environmental aspects in the area, including Yusufeli district.
For example, in a master’s thesis investigating the effects of build-
ing the Borçka and Deriner Dams on land use change and land deg-
radation, it was estimated that 546 ha of agricultural land along
the Murgul, Borçka, Ardanuç and central districts of Artvin city is
expected to be inundated by the reservoirs of these two dams (To-
ker, 2010).
Finally, it was determined that agricultural activities are carried
out in 13.31% (690 ha) of the land that is currently unsuitable and
in 5.14% (416.44 ha) of the land that is permanently unsuitable for
 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82
Fig. 8. Suitability map obtained after removing reservoir.
agriculture. By examining the physical properties of these lands, it
was determined that, on approximately 87%, the slope is higher
than 30%; 72% is included in class VII areas (27% class VIII); approx-
imately 59% has very shallow soil depth (33% litosolic); and, in
approximately 66%, the erosion degree is severe (very severe in
33%). It is known that the locals use soil for their own means, to
build terraces and carry out agricultural production on lands that
are unsuitable for agriculture.
4. Conclusions
The aim of this study was to determine the present agricultural
lands in Yusufeli district that will be left under the reservoirs of the
Artvin, Yusufeli and Deriner Dams and to find alternative areas
suitable for agriculture using GIS and AHP methods. The study
was conducted using 9 criteria reflecting the topographical proper-
ties and soil structure of the area. At the end of the assessment, it
was estimated that part of merely 8% of the study area is suitable
for agricultural production. It was determined that the low rate of
land which is suitable for agriculture originates from two main rea-
sons. The first and the most effective one is the fact that a signifi-
cant part of the land suitable for agriculture coincides with forest
and pasture areas. In order to preserve the forest and pasture areas,
the related regulation in force in Turkey prohibits the use of forest
and pasture areas for any other purpose, including agricultural pro-
duction. For this reason, these lands that cover approximately 85%
of the study area were excluded from the assessment. The second
reason is a sum of the problems originating from the geomorpho-
logical properties of the study area. Except for the slightly acidic
properties, the brown forest soils constituting the dominant great
soil group in the study area do not have any negative properties
81
against agricultural production. In addition, also the alluvial soils
present in the area are suitable for agricultural production. Fur-
thermore, the current climate and the presently cultivated agricul-
tural products indicate that the agricultural production potential of
Yusufeli is quite high. However, the problems caused by geomor-
phological status such as the high degree of slope, negatively affect
the degree of erosion, the rate of bare rocks, soil depth and land use
capability class. All these problems resulted in a very low amount
of land in the study area being found suitable for agricultural
production.
References
Akbulak, C., 2010. Analitik Hiyerarsßi Süreci ve Coğrafi Bilgi Sistemleri ile Yukarı Kara
Menderes Havzası’nın Arazi Kullanımı Uygunluk Analizi. Uluslararası Insan _
Bilimleri Dergisi 7 (2), 557–576.
Al-Shalabi, M.A., Mansor, S.B., Ahmed, N.B., Shiriff, R., 2006. GIS based multicriteria
approaches to housing site suitability assessment. In: XXIII FIG Congress,
October 8–13, Germany.
Amiri, F., Shariff, A.R.B.M., 2012. Application of geographic information systems in
landuse suitability evaluation for beekeeping: a case study of Vahregan
watershed (Iran). African Journal of Agricultural Research 7 (1), 89–97.
_ 2006. Toprak Olusßumu, Sınıflandırılması ve Coğrafyası. Meta Basım
Atalay, I.,
_
Matbaacılık, Izmir.
Aytekin, O., Aktasß, A., 2010. Yusufeli’de Yapılmaları Planlanan Hidro-Elektrik
Barajların Kültür Varlıklarına Olan Etkileri Üzerine Ön Değerlendirme,
Geçmisßten Geleceğe Yusufeli Sempozyumu, 10–12 Haziran. Yusufeli, Artvin.
Bandyopadhyay, S., Jaiswal, R.K., Hegde, V.S., Jayaraman, V., 2009. Assessment of
land suitability potentials for agriculture using a remote sensing and GIS based
approach. International Journal of Remote Sensing 30 (4), 879–895.
Demir, M., Demircioğlu Yıldız, N., Bulut, Y., Yılmaz, S., Özer, S., 2011. Alan Kullanım
Planlamasında Potansiyel Tarım Alanlarının Ölçütlerinin Coğrafi Bilgi Sistemleri
_
(CBS) Yöntemi ile Belirlenmesi (Ispir Örneği). Iğdır Üniversitesi Fen Bilimleri
Enstitüsü Dergisi 1 (3), 77–86.
Duc, T.T., 2006. Using GIS and AHP technique for land-use suitability analysis. In:
International Symposium on Geoinformatics for Spatial Infrastructure
82 H. Akıncı et al. / Computers and Electronics in Agriculture 97 (2013) 71–82
Development in Earth and Allied Sciences (GIS-IDEAS 2006), 9–11 November,
Ho Chi Minh City, Vietnam.
FAO, 1976. A Framework for Land Evaluation, Soils Bulletin 32. FAO and Agriculture
Organization of the United Nations, Rome.
Feizizadeh, B., Blaschke, T., 2012. Land suitability analysis for Tabriz County, Iran: a
multi-criteria evaluation approach using GIS. Journal of Environmental
Planning and Management 2012, 1–23.
Fu, Z., Li, Z., Zai, C., Shi, Z., Xu, Q., Wang, X., 2011. Soil thickness effect on
hydrological and erosion characteristics under sloping lands: a
hydropedological perspective. Geoderma 167–168, 41–53.
Gessler, P.E., Chadwick, O.A., Chamran, F., Althouse, L., Holmes, K., 2000. Modeling
soil-landscape and ecosystem properties using terrain attributes. Soil Science
Society of America Journal 64, 2046–2056.
_
Gök, Y., 2010. Tarihi Süreç Içerisinde Yusufeli Nüfusu, Geçmisßten Geleceğe Yusufeli
Sempozyumu, 10–12 Haziran. Yusufeli, Artvin.
Kalogirou, S., 2002. Expert systems and GIS: an application of land suitability
evaluation. Computers, Environment and Urban Systems 26 (2002), 89–112.
_
Koday, Z., Erhan, K., 2010. Yusufeli Ilçesinin _
Idari Coğrafya Analizi. Atatürk
Üniversitesi Sosyal Bilimler Enstitüsü Dergisi 14 (2), 1–11.
Koulouri, M., Giourga, C., 2007. Land abandonment and slope gradient as key factors
of soil erosion in Mediterranean terraced lands. Catena 69 (3), 274–281.
Lobo, D., Lozano, Z., Delgado, F., 2005. Water erosion risk assessment and impact on
productivity of a Venezuelan soil. Catena 64 (2–3), 297–306.
Malczewski, J., 1999. GIS and Multicriteria Decision Analysis. John Wiley and Sons,
New York.
Mendas, A., Delali, A., 2012. Integration of MultiCriteria Decision Analysis in GIS to
develop land suitability for agriculture: application to durum wheat cultivation
in the region of Mleta in Algeria. Computers and Electronics in Agriculture 83,
117–126.
Mokarram, M., Aminzadeh, F., 2010. GIS-based multicriteria land suitability
evaluation using ordered weight averaging with fuzzy quantifier: a case study
in Shavur Plain, Iran. The International Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences 38 (II), 508–512.
Mustafa, A.A., Singh, M., Sahoo, R.N., Ahmed, N., Khanna, M., Sarangi, A., Mishra, A.K.,
2011. Land suitability analysis for different crops: a multi criteria decision
making approach using remote sensing and GIS. Researcher 3 (12), 61–84.
Öztürk, D., Batuk, F., 2007. Çok Sayıda Kriter ile Karar Vermede Kriter Ağırlıkları.
Yıldız Teknik Üniversitesi Sigma Mühendislik ve Fen Bilimleri Dergisi 25 (1),
86–98.
Öztürk, D., Batuk, F., 2010. Konumsal Karar Problemlerinde Analitik Hiyerarsßi
Yönteminin Kullanılması. Yıldız Teknik Üniversitesi Sigma Mühendislik ve Fen
Bilimleri Dergisi 28, 124–137.
Parakash, T.N., 2003. Land Suitability Analysis for Agricultural Crops: A Fuzzy
Multicriteria Decision Making Approach. MSc Thesis, The International Institute
for Geo-information Science and Earth Observation (ITC), Enschede, The
Netherlands.
Perveen, M.F., Nagasawa, R., Uddin, M.I., Delowar, H.K.M., 2007. Crop-land
suitability analysis using a multicriteria evaluation & GIS approach. In: 5th
International Symposium on Digital Earth (ISDE5), June 5–9, University of
California, Berkeley, USA.
Saaty, T.L., 1977. A scaling method for priorities in hierarchical structures. Journal of
Mathematical Psychology 15, 57–68.
Saaty, T.L., 1980. The Analytic Hierarchy Process: Planning, Priority Setting,
Resource Allocation. McGraw-Hill International, New York, NY, USA.
Saaty, T.L., 1990. An exposition of the AHP in reply to the paper ‘remarks on the
analytic hierarchy process’. Management Science 36, 259–268.
Saaty, T.L., 1994. Fundamentals of Decision Making and Priority Theory with the
AHP. RWS Publications, Pittsburgh, PA, USA.
Saaty, T.L., 2004. Mathematical Methods of Operations Research. Dover
Publications, Mineola, pp. 415–447.
Soil Atlas of Europe, 2005. European Soil Bureau Network, European Commission,
Office for Official Publications of the European Communities, L-2995
Luxembourg.
Toker, E., 2010. Borçka ve Deriner Barajlarının Çoruh Havzasında Neden Olduğu
_
Arazi Kullanım Değisßiminin ve Arazi Tahribatının Irdelenmesi, Yüksek Lisans
Tezi, Artvin Çoruh Üniversitesi Fen Bilimleri Enstitüsü. Orman Mühendisliği
Anabilim Dalı, Artvin.
Tombusß, F.E., 2005. Uzaktan Algılama ve Coğrafi Bilgi Sistemleri Kullanılarak
Erozyon Risk Belirlemesine Yeni Bir Yaklasßım, Çorum Ili _ Örneği, Yüksek Lisans
Tezi, Anadolu Üniversitesi, Fen Bilimleri Enstitüsü. Uzaktan Algılama ve Coğrafi
Bilgi Sistemleri Anabilim Dalı, Eskisßehir.
TRGM, 2008. Toprak ve Arazi Sınıflaması Standartları Teknik Talimatı ve Ilgili _
Mevzuat, <http://www.tarim.gov.tr/Documents/Mevzuat/Talimatlar/
ToprakAraziSiniflamasiStandartlariTeknikTalimativeIlgiliMevzuat_yeni.pdf>
(accessed 03.06.13).
Triantaphyllou, E., Mann, S.H., 1995. Using the analytic hierarchy process for
decision making in engineering applications: some challenges. International
Journal of Industrial Engineering: Applications and Practice 2 (1), 35–44.
USDA, 2008. The United States Department of Agriculture (USDA) National Soil
Survey Handbook, Part 622. <http://soils.usda.gov/technical/handbook/
contents/part622.html> (accessed 03.06.13).
Wang, F., 1994. The use of artificial neural networks in a geographical information
system for agricultural land-suitability assessment. Environment and Planning
A 26, 265–284.
Yalçın, G., Eken, G., 2006. Türkiye’nin Baraj Politikası ve Önemli Doğa Alanları -
Doğa Derneği Kurumsal Görüsßü, TMMOB Su Politikaları Kongresi Bildiriler
Kitabı, 21–23 Mart. Ankara, Cilt 1, 250–265.
Zengin, M., Yılmaz, S., 2008. Ardahan Kura Nehri ve Yakın Çevresi Alan
Kullanımlarının Belirlenmesi ve Optimal Alan Kullanım Önerileri. Atatürk
Üniversitesi Ziraat Fakültesi Dergisi 39 (1), 43–54.